α-Fluorinated cyclic amidophosphite ligands. Their synthesis, Rh complexes and catalytic activity in the hydroformylation of styrene

Article abstract of DOI:10.1016/j.jorganchem.2006.08.098

The synthetic approaches to cyclic phosphite and amido(diamido)phosphite ligands bearing the residues of electron withdrawing perfluorinated tails at the β-position to the phosphorus atom have been elaborated. Catalytic systems based on rhodium complexes

Full text of DOI:10.1016/j.jorganchem.2006.08.098

Journal of Organometallic Chemistry 691 (2006) 5547–5559
www.elsevier.com/locate/jorganchem
a-Fluorinated cyclic amidophosphite ligands. Their synthesis,
Rh complexes and catalytic activity in the hydroformylation of styrene
a
a,
a
a
Oleg Artyushin , Irina Odinets *, Evgenii Goryunov , Ivan Fedyanin ,
a
a
c
Konstantin Lyssenko , Tatyana Mastryukova , Gerd-Volker Roschenthaler ,
¨
b
d
e
´
´
´
´
´
Tamas Kegl , Gyo¨rgy Keglevich , Laszlo Kollar
a
A.N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, Thiophosphorus Laboratory,
Vavilova Str., 28, 119991 Moscow, Russian Federation
b
Research Group for Petrochemistry of the Hungarian Academy of Sciences, H-8200 Veszpre´m, Hungary
c
Institut f€ur Anorganische & Physikalische Chemie, Universita¨t Bremen, Germany
Department of Organic Chemical Technology, Budapest University of Technology and Economics, 1521 Budapest, Hungary
d
e
Department of Inorganic Chemistry, University of Pe´cs and Research Group for Chemical Sensors of the Hungarian Academy of Sciences,
H-7624 Pe´cs, Hungary
Received 26 July 2006; received in revised form 24 August 2006; accepted 24 August 2006
Available online 15 September 2006
Abstract
The synthetic approaches to cyclic phosphite and amido(diamido)phosphite ligands bearing the residues of electron withdrawing
perfluorinated tails at the b-position to the phosphorus atom have been elaborated. Catalytic systems based on rhodium complexes
of these ligands formed in situ using Rh(CO)2(acac) as a catalytic precursor demonstrate high activity in the hydroformylation of styrene
along with good selectivity in respect to branched aldehyde. Quantum-chemical calculations proved that both the rate of the formation of
branched alkyl complex, as well as its reactivity are influenced by the steric and electronic parameters in the same manner.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Polyfluorinated cyclic phosphite and amidophosphite ligands; Rhodium complexes; Homogeneous catalysis; Hydroformylation
1. Introduction
molecule. This fact is essential in the synthesis of some
coordination compounds and in hydroformylations where
oxidation of phosphines deems a marked problem. Fur-
thermore, these ligands have pronounced p-acceptor prop-
erties allowing the stabilization of low oxidation state of
the complexing metals and an increase of its electrophilic-
ity. In general, the introduction of oxygen and/or nitrogen
heteroatoms into the first coordination sphere of the phos-
phorus atom presents a more effective way to adjust the
chemical stability, as well as the donor–acceptor and steric
properties of the ligand, as compared to the traditional
introduction of substituents into the phenyl substituents
of PPh2 moieties typical for most phosphines.
Nowadays phosphite or/and amidophosphite ligands
attract attention in metal catalysed processes. These com-
pounds possess a series of principal advantages in compar-
ison with r-donor phosphine ligands [1]. On one hand,
they are easily available, as most of them may be obtained
via direct one step phosphorylation from a variety of pre-
cursors including optically active ones; at the same time,
the synthesis of optically active phosphines is more compli-
cated. On the other hand, these compounds are more stable
against oxidation due to the absence of P–C bonds in the
In hydroformylation, being one of the most important
homogeneous catalytic industrial processes, small r-donor
ligands was found to retard the reaction while p-acceptor
*
Corresponding author. Tel.: +7 95 1359356; fax: +7 95 1355085.
E-mail address: odinets@ineos.ac.ru (I. Odinets).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.08.098

5548
O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
ligands, such as bulky phosphites [2] or phosphabenz-
enes [3] possessed high catalytic activity for both hydro-
formylation of internal and terminal olefins. In these
cases the enormous catalytic activity was attributed to a
monoligand–rhodium complex as a catalytically active
species.
activity in the hydroformylation of styrene as well as quan-
tum-chemical estimation of structure–activity relationship,
of novel cyclic amidophosphite ligands bearing polyfluori-
nated tails are reported.
2. Results and discussion
Taking into account that fluorinated ligands also find
application in the field of catalysis possessing high activity
and are of high interest in fluorous biphasic catalysis (FBC)
[4], recently [5], we elaborated the synthesis of non-sym-
metric benzoannulated cyclic phosphites 1 bearing the res-
idue of substituted benzylic alcohols containing fluorine
substituents in aliphatic and/or aromatic positions. In this
case, electron withdrawing polyfluorinated tails were used
to stabilize the phosphorus(III) benzylic esters which other-
wise would undergo a rather rapid decomposition [6].
The rhodium complexes of RhIII(L)(Cp*)Cl2 type or RhI
complexes formed in situ from Rh(CO)2(acac) precursor
and the corresponding ligand proved to be active as cata-
lysts in the hydroformylation of styrene.
2.1. Synthesis
In principle, two general methodologies can be used to
obtain (amido)phosphites. The more common route is
based on the so called ‘chloride methodology’, wherein
the corresponding pre-formed cyclic phosphorus(III) acid
chloride interacts with the corresponding alcohol in the
presence of a base (Scheme 1, path A). The treatment of
dichlorophosphites by bisnucleophiles means one variant
of this method (Scheme 1, path B). The other route, the
so-called ‘‘amide’’ procedure, is based on the interaction
of phosphorus(III) acid amides with the desired alcohols
(Scheme 1, path C). All the approaches shown in Scheme
1 were used in our work to find the optimum choice for
the synthesis of the target molecules.
Thus, to obtain the ligands 1 from easily available 2-
chloro-1,3,2-benzodioxaphosphole 2, the ‘chloride method’
(pathway A) was found to be the best choice [5]. The same
approach was used to synthesize their analogues bearing
either an alkyl chain or a diethoxyphosphoryl group
instead of the aromatic ring adjacent to trifluoromethyl
substituent (3) (Scheme 2). The original procedure [5a]
involved minor modifications (triethylamine was used
O
P OCH Ar
O
1
CX3
X=F, Ar=C6H5, 3-Cl-C6H4, 3-CF3C6H4;
X=H, Ar=C6F5
Both systems provided excellent hydroformylation
activities at 100 ꢁC (TOF values up to 2000) excluding
the ligand bearing the m-chlorine atom. At this tempera-
ture only moderate regioselectivities towards the branched
aldehyde (2-phenylpropanal) were, however, observed.
Using the RhI in situ systems, high regioselectivities were
observed already at 40 ꢁC, the activity of the catalyst
decreased, however, under such conditions and the com-
plete conversion took ca. 72 h.
Taking into account that many examples are known
where minor modifications in the structure of the ligand
resulted in a profound change in of the catalytic activity
and/or in the selectivity, it seemed to be interesting to esti-
mate the structure–activity relationship for such ligands
varying either the heteroatoms in the first coordination
sphere of the phosphorus or the substituents at the a-car-
bon atom. In this paper, the synthesis and the catalytic
X
HOR1/B:
P
Cl
A:
-B.HCl
Y
X
Y
Cl
Cl
HX
YH
P
OR1
OR1
P
B:
C:
X
Y
HOR1
P
NEt2
-Et2NH
X,Y=N(R) or O
Scheme 1.
O
O
P
HOCH(CX3)R / Et3N
P
Cl
OCH
CF3
R
O
O
2
3a,b
R=n-C6H13 (a), P(O)(OEt)2 (b)
Scheme 2.

O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
5549
instead of pyridine and the order of the addition of the
reagents was changed (see Section 3)).
(e.g. iPr, tBu, Ph) on the nitrogen atoms, as we failed to
obtain the starting cyclic diamidochlorophosphites 10
from phosphorus trichloride and the corresponding
N,N0-dialkyl(aryl)alkylene diamine (Scheme 5).
Even applying low temperatures (ꢀ50 ꢁC) and working
in diluted solutions, the desired interaction to afford 10
(n = 2,3) was accompanied by the formation of linear
aminochlorophosphites of type 11 and 12 of oligomeric
nature.
Then we switched for the preparation of amidophosph-
ites. As ortho-N-methylaminophenol can be prepared in
only ca. 10% yield [7], we used the ‘amide’ procedure (path
C) giving amidophosphites in high yields and purities,
applying N-diethylamidophosphite 4 as the starting mate-
rial for the synthesis of monoamidophosphites 5a–e
(Scheme 3).
Non-fused amidophosphites 8a–c and diamidophosph-
ites 9a–c bearing the N-methyl group(s) were synthesized
The yields of cyclic products 10 (n = 2, 3) estimated by
31P NMR spectra were less than 60% in the case of the sub-
strate with isopropyl group. Moreover, we failed to isolate
products 10 by distillation of the crude mixtures, as the lin-
ear products underwent disproportionation. The propor-
tion of cyclic chlorophosphites 10 decreased till 20% by
increasing the steric hindrance at the nitrogen atoms and
by increasing the length of the alkylene bridge in the start-
ing alkylene diamine.
Therefore, during our efforts to obtain cyclic
diamidophosphites of the above type, dichlorophosphites
13a,b were obtained first by the reaction of PCl3 and the
corresponding fluorinated alcohols followed by the cycliza-
tion with N0,N-dialkyl(aryl)alkylene diamines (Scheme 6).
It should be noted that due to the weak nucleophilicity of
fluorinated alcohols, the yields of intermediates 13
obtained by typical procedure were only ca. 50%.The use
of catalytic amount of 4-dimethylaminopyridine (DMAP)
from 2-chloro-3-methyl-1,3,2-oxazaphospholidine
6 or
2-chloro-1,3-dimethyl-1,3,2-diazaphospholidine 7, respec-
tively (Scheme 4). The compounds (8a–c and 9a–c) were
isolated in analytically pure form after the work-up proce-
dure in yields of 82–98%. Note that as amidophosphites 5
and 8 possess a stereogenic phosphorus atom along with
the stereogenic carbon atom in the side-chain, they were
obtained as a mixture of two diastereomers observed in
the 31P NMR spectra as two singlets. The diastereomer A
having upfield chemical shift in the 31P NMR spectra was
usually the major one and the ratio of isomers ranged
1
between 1.3 and 1.5:1. In the H and 13C NMR spectra
two series of signals corresponding to A and B diastereo-
mers were observed (see Section 3.2).
This procedure could not be used for the preparation of
diamidophosphites bearing sterically hindered substituents
Me
Me
Me
N
N
NH
Et2NPCl2
HOCH(CX3)R
P
OCH
CX3
R
P
NEt2
160˚C, 1 h
O
O
OH
4
5a-e
X=F, R=C6H5 (5a), 3-CH3OC6H4 (5b); 3-CF3C6H4 (5c); n-C6H13 (5d)
X=H, R=C6F5 (5e)
Scheme 3.
Me
N
P
P
OCH(CX3)R
8a-c
O
HOCH(CX3)R/Et3N
Z
Y
P
Cl
Me
N
6,7
OCH(CX3)R
Me
N
Me
N
N
Z
Y
Me
=
P
P
(6);
P
(7);
9a-c
O
N
Me
X=F, R=3-CF3C6H4 (8a), P(O)(OEt)2 (8b); X=H, R=C6F5 (8c)
X=F, R=C6H5 (9a), P(O)(OEt)2 (9b); n-C6H13 (9c);
Scheme 4.

5550
O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
R'
R'
N
N(CH2)n
(CH2)n
PCl3 + R'NH(CH2)nNHR'
PCl
Cl2PN(CH2)nNPCl2
P
+
+
m
R'
R'
Cl
N
R=i-Pr, t-Bu, Ph
n=2, 3
11
12
R'
10
Scheme 5.
R1
N
R1HN
NHR1
HOCH(CX3)R
PCl3
(CH2)n
Et3N
Cl2POCH(CX3)R
(CH2)n
POCH(CX3)R
cat. DMAP
13a,b
N
R1
14a-g
X=F, R=C6H5 (13a); X=H, R=C6F5 (13b)
R1=i-Pr: X=F, R=C6H5, n=2 (14a), n=3 (14b); X=H, R=C6F5, n=2 (14c)
R1=t-Bu: X=F, R=C6H5, n=2 (14d); X=H, R=C6F5, n=2 (14e);
R1=Ph: X=F, R=C6H5, n=2 (14f); X=H, R=C6F5, n=2 (14g)
Scheme 6.
allowed, however, both the increase of the yield up to 77%
and the accomplishment of the substitution at room tem-
perature [8]. Applying this approach, diamidophosphites
14 could be obtained in good yields (83–87%, based on
13a,b or ca. 60% on the basis of PCl3).
299.1ꢁ) while in contrary the N(2) arom is planar (sum
of bond angles is 359.8ꢁ). The Ph rings at both nitrogen
atoms of the cycle are almost coplanar with the base of
envelope with torsion angles C(12)C(11)N(1)P(1) and
C(22)C(17)N(2)P(1) equal to 22.9ꢁ and 8.3ꢁ, respectively.
The phenyl ring C(3)–C(8) in alkoxy group is situated
directly above the C(17)–C(22) one with the dihedral
angle between the planes of aromatic cycles equal to
20.5ꢁ and the shortest C(11)ꢁ ꢁ ꢁC(20) contact equal to
All compounds were characterized by elemental analy-
sis or MS data, as well as NMR spectroscopy. The ste-
reostructure of ligand 14f was evaluated by single
crystal X-ray analysis. According to the data obtained,
diamidophosphite 14f crystallises as a racemate. The
five-membered 1,3,2-diazaphospholidine cycle is charac-
terized by the envelope conformation with the deviation
˚
3.430(2) A. Such mutual disposition of these phenyl rings
almost screens the acidic hydrogen atom at the chiral
C(1) atom and as the consequence, it does not participate
in any shortened intermolecular contacts in the crystal of
14f.
˚
of the C(10) atom by 0.32 A (Fig. 1). The N(1) and
P(1) atoms are pyramidal (sum of bond angles 356ꢁ and
˚
Fig. 1. Perspective view of 14f with representation of atoms by the thermal ellipsoids at 50% probability level. The selected bond lengths (A): P(1)–O(1)
1.657(1), P(1)–N(2) 1.693(2), P(1)–N(1) 1.701(2), O(1)–C(1) 1.429(2), N(1)–C(11) 1.410(2), N(1)–C(9) 1.474(2), N(2)–C(17) 1.409(2), N(2)–C(10) 1.461(2);
bond angles (ꢁ): O(1)–P(1)–N(2) 105.04(8), O(1)–P(1)–N(1) 103.58(8), N(2)–P(1)–N(1) 90.51(8), C(1)–O(1)–P(1) 122.83(12), C(11)–N(1)–C(9) 118.94(16),
C(11)–N(1)–P(1) 121.52(13), C(9)–N(1)–P(1) 115.61(12), C(17)–N(2)–C(10) 120.82(15), C(17)–N(2)–P(1) 123.26(14), C(10)–N(2)–P(1) 115.70(12).

O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
5551
Z
Y
racemate. Therefore, the complex 15a presents a conglom-
erate, which undergoes crystal induced spontaneous resolu-
tion to give enantiopure crystals. It should be noted that
for (p-pentamethylcyclopentadienyl)(phosphite)rhodium
dichlorides LRhCp*Cl2 the spontaneous resolution was
also observed by us earlier in the case of the complex
formed by 2,2,2-trifluoro-1-phenylethyl diethylphosphinite
(L = Et2POCH(CF3)C6H5) [4b].
Z
Y
OR
P
OR + 1/2 [RhCp*Cl2]2
P
RhCp*Cl2
5c, 9a,b
15a-c
L=5c (15a); 9a (15b); 9b (15c)
Scheme 7.
Similar to phosphites 1, the mono- and the
diamidophosphites obtained were suitable P-ligands to give
the stable Rh(III) complexes as illustrated by the reaction
of compounds (5c, 9a,b) with dimeric (p-pentamethylcyclo-
pentadienyl)rhodium(III) dichloride (Scheme 7).
2.2. Hydroformylation in the presence of rhodium—fluorine-
containing phosphite and amidophosphite in situ catalysts
Styrene as a model substrate was allowed to react with
CO/H2 (1/1) at 40 ꢁC or 100 ꢁC at 100 bar in the presence
of rhodium–L in situ catalysts formed from Rh(CO)2(acac)
and the corresponding phosphite or amidophosphite ligand
(where L represents 3a, 3b, 5c, 5d, 8a, 8c, 14a and 14b; for
structural formulae see Scheme 2,3,4 and 6) over 2 h. The
above phosphorus ligands differ significantly both in elec-
tronic and in steric properties (see Section 2.3).
In addition to the formation of the two formyl regio-
isomers, 2-phenyl-propanal (I) and 3-phenyl-propanal
(II), that of ethylbenzene (III) arising from hydrogenation
was also expected (Scheme 8).
According to the X-ray diffraction analysis of the
complex 15a formed by substituted 2,3-dihydro-1,3,2-ben-
zoxazaphosphole 5c (Fig. 2), the principal geometrical
parameters are similar to those in its analogue formed by
the similar ligand excluding of dioxaphospholene ring
instead of 3-methyl-1,3,2-benzoxazaphospholene ring [5a]
with slight decrease of the Rh(1)–P(1) bond up to
˚
˚
2.231(2) A in comparison to 2.259(1) A published previ-
ously. The phosphorus atom is characterized by the dis-
torted P-pyramid with angles for OPO and OPN reduced
up to 92.6(4)–106.7(4)ꢁ and with increase of the
Rh(1)P(1)N(1) angle (124.0(3)ꢁ). Furthermore, the X-ray
analysis have revealed that 15a crystallises in the chiral
space group C2 (the Flack parameter is equal to 0.02(6))
in contrast to the above mentioned complex with dioxa-
phospholene ring published in [4a] which crystallized as a
CO/H2
PhCH CH2
PhCH(CHO)CH3 + PhCH2CH2CHO + PhCH2CH3
I
II
III
Scheme 8.
Fig. 2. Perspective view of 15a with representation of atoms by the thermal ellipsoids at 50% probability level. The fluorine atoms of the disordered CF3
˚
group are shown by balls. The selected bond lengths (A): Rh(1)–P(1) 2.231(2), Rh(1)–Cl(2) 2.393(3), Rh(1)–Cl(1) 2.408(3), Rh(1)–Cp*cent 1.815(9), P(1)–
O(1) 1.616(6), P(1)–O(2) 1.624(7), P(1)–N(1) 1.688(9); bond angles (ꢁ): P(1)–Rh(1)–Cl(2) 89.46(12), P(1)–Rh(1)–Cl(1) 89.08(10), Cl(2)–Rh(1)–Cl(1)
91.40(10), O(1)–P(1)–O(2) 102.2(4), O(1)–P(1)–N(1) 106.7(4), O(2)–P(1)–N(1) 92.6(4), O(1)–P(1)–Rh(1) 110.3(2), O(2)–P(1)–Rh(1) 118.0(3), N(1)–P(1)–
Rh(1) 124.0(3).

5552
O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
Table 1
ophosphites are slightly higher or lower than 90%, those
obtained with diamidophosphites (14a and 14b) are typi-
cally about 80%.
Although the above hydroformylations were not carried
out with the aim of clearing-up some mechanistic details,
according to a generally accepted mechanism (Scheme 9),
some conclusions could be drawn on the basis of the cata-
lytic results.
Hydroformylation of styrene in the presence of in situ catalyst formed
from Rh(acac)(CO)2 and La
c
d
L
Conv.b (%)
RC (%)
RRB (%)
3a
3b
5c
5d
8a
8c
14a
14b
a
99
>99.5
98
>99.5
88
97
86
61
99
99
>99.5
>99.9
>99.9
99
87
90
91
90
87
89
81
77
As expected, the RhI precursor provided the
RhH(CO)n(L)m (n,m = 1,2) species by facile substitution
of the acac ligand under ‘hydroformylation conditions’.
Furthermore, the RhH(CO)(L)2 key-intermediate, as well
as the corresponding Rh(alkyl)(CO)(L)2 intermediate,
formed by styrene insertion, are easily available in the
in situ system and could be formed even at such a relatively
low temperature, as 70 ꢁC. It is worth noting that the for-
mation of the branched alkyl complex (Rh(1-phenyl-
ethyl)(CO)(L)2 intermediate) is favoured in the presence
of all ligands. From the two major reaction pathways based
on this alkyl-intermediate, i.e. carbon monoxide insertion
followed by hydrogenolysis, as well as direct hydrogenoly-
sis of the Rh-alkyls, the previous one is highly favoured. As
a consequence of this, almost the exclusive formation of the
aldehydes (I and II) and only traces of the hydrogenated
product (III) could be observed in our experiments.
In conclusion, some representative examples of novel
fluorinated phosphite and amidophosphite P-ligands were
tested in the homogeneous catalytic hydroformylation of
styrene. The systems demonstrated high catalytic activity,
excellent chemoselectivity towards hydroformylation and
moderate to high (up to 91%) regioselectivity towards the
branched aldehyde. The activity of compounds with fused
1,3,2-benzoxaphospholene or 1,3,2-benzazaphospholene
cycles was comparable with that of the most active rho-
dium–monophosphine catalysts.
99
>99.9
Reaction conditions: p = 100 bar (CO:H2 = 1:1), reaction tempera-
ture = 70 ꢁC, r. time = 2 h, Rh:L:styrene = 1:2:500, solvent: toluene.
b
Determined by GC.
Chemoselectivity = (moles of I + moles of II)/(moles of I + moles of
c
II + moles of III) · 100.
d
Regioselectivity = (moles of I)/(moles of I + moles of II) · 100.
All the in situ catalysts were active at 70 ꢁC under the
given conditions (Table 1). Practically complete conver-
sions have been obtained in 2 h with tris(phosphite) type
ligands (3a, 3b) and amidophosphites 5c and 5d having
fused 1,3,2-benzoxaphospholene or 1,3,2-benzazaphospho-
lene cycles respectively. Decreased activity was observed
with the monocyclic ligands 8a, 14a,b while the ligand 8c
having perfluorophenyl substituent instead of trifluro-
methyl group in 8a still demonstrated 97% conversion.
The lowest activity was observed with the diamidophosph-
ite ligand 14b with six-membered 1,3,2-diazaphosphorine
cycle.
The reaction was highly chemoselective towards the for-
mation of aldehydes. Excellent chemoselectivities (higher
than 99%, in some cases higher than 99.9%) were obtained
in all cases. The regioselectivities towards the branched
aldehyde regioisomer (I) fall in the range of 77–91%, i.e.
the branched aldehyde (I) predominated over the linear
one (II). Although the range of regioselectivity was rather
narrow, two groups of ligands can be distinguished: the
regioselectivities obtained with phosphites and amid-
2.3. Theoretical studies
Among the numerous examples of phosphite and amid-
ophosphite ligands discussed in this study some have been
selected in order to find some correlation between the elec-
tronic and steric properties and the activity and regioselec-
tivity in the rhodium-catalyzed hydroformylation of
styrene.
The lowest energy conformers of the selected ligands
3a,b, 5c,d, 8a,c, 14a,b optimized at PBE level of theory
are depicted in Fig. 3, the electronic and steric data are
summarised in Table 2. As the influence of steric effects
on organometallic compounds are generally discussed
using Tolman’s cone angle concept [9], we determined the
cone angle (H) on all the lowest energy conformers of each
ligand. Some electronic properties characteristic for coordi-
nation to transition metals were determined using the nat-
ural bond orbital (NBO) method [10]. The natural charge
(QP) describes the charge on the phosphorus atom obtained
by natural population analysis, i.e. the nuclear charge
minus the total electron population. The natural electron
Ph
Ph
L2(CO)[Rh]-H
CHO
H2
+
Ph
CHO
Ph
L2(CO)[Rh]
H
Ph
CO
L2(CO)[Rh]
L2(CO)[Rh]
+
Ph
CO
Ph
L2(CO)[Rh]
+
Ph
L2(CO)[Rh]
CO
Scheme 9.

O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
5553
˚
Fig. 3. PBE/6-31G*-optimized structures of ligands selected for hydroformylation experiments. P–O and P–N bond lengths are given in Angstrom.
show significant difference. The n-hexyl group greatly
increases, however, the cone angle of the ligands (see 5d
vs. 5c and 3a vs. 3b) thus the slight decrease of regioselec-
tivity may be attributed to steric effect.
The influence of the fused phenyl group may be followed
comparing ligands 5c and 8a. The more rigid structure of
the benzooxazaphospholidinyl group results in slightly
elongated P–N and P–O bond length, a lone pair on phos-
phorus with more s character and also higher catalytic
activity and selectivity to the branched aldehyde.
Table 2
NBO properties and Tolman’s cone angles of ligands selected for catalytic
experiments
NEC (valence)b
NHOP,lp
H (o)
a
c
Ligand
QP
3a
3b
5c
5d
8a
8c
14a
14b
a
1.505
1.522
1.447
1.436
1.451
1.451
1.344
1.369
3S(1.59)3p(1.83)
3S(1.58)3p(1.81)
3S(1.55)3p(1.93)
3S(1.55)3p(1.94)
3S(1.53)3p(1.94)
3S(1.52)3p(1.95)
3S(1.50)3p(2.09)
3S(1.47)3p(2.10)
sp0.39
sp0.40
sp0.46
sp0.45
sp0.49
sp0.49
sp0.55
sp0.60
133
112
132
155
130
134
151
156
The ring size and hence the ring strain of the cyclic
diamidophosphites 14a and 14b influences significantly
the electronic structure around the phosphorus atom. Both
the catalytic activity and the selectivity are dropped when
n = 3. The more positively charged phosphorus and the
lone pair with more p character may be attributed to the
larger bite angle and shorter P–N bonds. The P–N bonds
of 14b become more polarized, as the natural charge on
nitrogen atoms decrease from ꢀ0.783 and ꢀ0.798 to
ꢀ0.787 and ꢀ0.802, respectively.
NBO charge on the phosphorus atom.
b
c
Natural electronic configuration on phosphorus.
Natural bond orbital hybridization on the phosphorus lone pair.
configuration (NEC) gives the effective valence electron
configuration for the phosphorus atom. The lone pair of
the phosphorus is composed entirely of a single normalized
natural hybrid orbital (NHO).
The role of the cone angle was mainly examined for the
ligand pairs 3a–3b and 5c–5d, respectively. Although the
electron density is slightly more depleted in case of 5c
and 3b as compared to 5d and 3a, respectively, the electron
configuration and the p character of the lone pairs do not
In accordance with the simultaneous change of catalytic
activity and branched selectivity, it may be concluded that
both the rate of the formation of branched alkyl complex
as well as its reactivity are influenced by the steric and elec-
tronic parameters in the same manner.

5554
O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
To conclude, the application of a combination of ‘chlo-
(0.35 g, 2 mmol) in 30 mL of diethyl ether under argon
and the mixture was stirred over additional 5 min. Then
the solution of 2 mmol of the corresponding alcohol in
30 mL of diethyl ether was added. The mixture was allowed
to warm up to ambient temperature and stirred for addi-
tional 3 h. The precipitate was filtered off under argon,
washed with Et2O (2 · 30 mL) and discarded. The filtrate
was evaporated in vacuo. The residue was treated with
30 mL of pentane and kept at 0 ꢁC overnight. The clear
solution was decanted from the precipitate and evaporated
under reduced pressure followed by drying (20 ꢁC,
0.1 mmHg, 2 h) to afford the desired unsymmetric tris-
phosphites 3a,b as colourless oils.
ride’ and ‘amide’ methodologies gives the possibility to
obtain a variety of asymmetric cyclic phosphites and
amidophosphites bearing the residue of weak nucleophilic
fluorinated alcohols in high yields. Catalytic systems
based on rhodium complexes of these ligands formed
in situ demonstrate high activity in the hydroformylation
of styrene. According to the ‘structure–activity’ relation-
ship the more rigid structure of the ligand and the pres-
ence of the nitrogen atom at the phosphorus result in
higher catalytic activity and selectivity to the branched
aldehyde.
3. Experimental
3.2.2. 2-{[1-(Trifluoromethyl)heptyl]oxy}-1,3,2-
benzodioxaphosphole 3a
3.1. General
Yield 87%, 31P NMR: 133.5; 19F NMR: ꢀ78.7
3
The NMR spectra were recorded on a ‘Bruker-DPX-
200’ and ‘Bruker AMX-400’ spectrometer in CDCl3 solu-
tions using residual proton signals or the characteristic
13C chemical shift of the deuterated solvent as an internal
standard (for 1H and 13C NMR, respectively) and 85%
H3PO4 as an external standard (for 31P–{1H} NMR).
The 19F chemical shifts were determined with CFCl3 as
an external standard. The mass-spectra (EI) were obtained
on a Varian MAT-311A spectrometer at 70 eV.
All reactions were conducted under an inert atmosphere
of dry argon by using Schlenk glassware and vacuum line
techniques. Solvents were freshly distilled from standard
drying agents. Toluene was distilled under argon from
sodium in the presence of benzophenone. Styrene was
freshly distilled before use.
The starting 1-(2,3,4,5,6-pentafluorophenyl)-ethanol
was purchased from Aldrich and used in synthesis without
purification. a-Trifluoromethylbenzyl alcohols were syn-
thesized by the reduction of the corresponding ketones
according the procedure developed by us earlier [11].
Diethyl 2,2,2-trifluoro-1-hydroxyethylphosphonate was
synthesized by the interaction of trifluoroacetic acid ethyl
ester and diethylphosphite by the procedure described in
(appeared t, JF–H = 4JP–F = 7.5 Hz); 1H NMR: 4.2–4.0
3
(m, CH, 1H); 0.92 (t, JH–H = 6 Hz, CH3, 3H); 1.62 (q,
3JH–H = 6 Hz, ICCH2, 2H); 1.4–1.2 (m, –(CH2)4–, 8H);
7.2–7.0 (m, C6H4, 4H). 13C NMR: 13.9 (CH3); 22.4, 24.3;
3
1
28.5; 28.9; 31.36; 123.5 (qd, CF3, JP–C = 3.6 Hz, JC–F
=
2
2
276.1 Hz); 72.6 (dq, CH, JC–F = 31.9 Hz, JP–C = 6.2);
3
3
111.1 (d, JP–C = 12.7 Hz), 112.3 (d, JP–C = 1.9 Hz),
4
2
122.0; 123.0 (d, JP–C = 4.4 Hz), 143.8 (d, CArOP, JP–C
=
2
3.5 Hz), 144.8 (d, CArOP, JP–C = 7.1 Hz). EI-MS, m/z
323 (M+H)+.
3.2.3. Diethyl 1-(1,3,2-benzodioxaphosphol-2-yloxy)-2,2,2-
trifluoroethylphosphonate 3b
Yield 91%, 31P NMR: 137.3 (br.s, PIII), 10.7 (m, PIV); 19
F
3
4
NMR: ꢀ72.15 (dd, JF–H = 7.5 Hz, JP–F = 13 Hz); 1H
NMR: 4.6–4.4 (m, CH, 1H); 4.4–4.1 (m, OCH2, 4H); 1.35
(t, JH–H = 6 Hz, CH3, 6H); 7.25–7.0 (m, C6H4, 4H). 13C
3
3
NMR; 15.8 and 15.9 (both d, CH3, JP–C = 5.1 Hz); 64.0
2
and 64.1 (both d, POCH2, JP–C = 21.1 Hz); 68.2 (dqd,
2
2
1JP(IV)-C = 168.0 Hz, JP(III)-C = 8.8 Hz, JC–F = 34.3 Hz);
3
2
112.4 (d, JP–C = 2.1); 121.6 (qdd, JP(IV)-C = 5.1 Hz,
3JP(III)-C = 5.1 Hz, JC–F = 280.0 Hz); 123.0; 144.1 (d,
1
2
2
CArOP, JP–C = 6.0 Hz), 144.2 (d, CArOP, JP–C
=
Ref. [12]. 2-Chloro-1,3,2-benzodioxaphosphole
2
was
5.8 Hz). Anal. Calc. for C12H15F3O6P2: C, 38.52, H, 4.04;
P, 16.55. Found: C, 38.46; H, 4.14; P, 16.39%.
obtained by a known procedure [13]. 2-Chloro-3-methyl-
1,3,2-oxazaphospholidine 6 was obtained from phosphorus
trichloride and 2-N-(methylamino)ethanol as it was
described [14], but with minor modifications (solvent:
50 mL of diethyl ether, for 4.81 g (0.035 mol) of PCl3, 2 h
at ꢀ50 ꢁC ! 2 h at 20 ꢁC) to increase the yield up to 51%
(vs. 24%). 2-Chloro-1,3-dimethyl-1,3,2-diazaphospholidine
7 was prepared according to Ref. [15]. Dichlorophosphites
13a,b were obtained according to the catalytical procedure
elaborated by us earlier [8].
3.2.4. Amidophosphites 5a–e
3.2.4.1.
N,N-Diethyl-3-methyl-1,3,2-benzoxazaphosphol-
2(3H)-amine 4. To a cooled (ꢀ10 ꢁC) solution of diet-
hylamidodichlorophosphite (4.39 g, 0.0252 mol) in diethyl
ether (100 mL) the mixture of N-methyl-o-aminophenol
(3.1 g, 0.0252 mol) and triethylamine (5.09 g. 0.0504 mol)
in 100 mL of Et2O was added. The mixture was allowed
to warm up to room temperature and stirred for 2 h. Then,
the precipitate was filtered off, the solvent was evaporated
and the residue was distilled in vacuum to give 3.58 g
(63%) of amidophosphite 4, b.p. 160–162 ꢁC (15 mmHg);
3.2. Synthesis of the P-ligands
1
3
3.2.1. Phosphites 3a,b
Triethylamine (0.2 g, 2 mmol) was added to a stirred
and cooled (ꢀ5 to 0 ꢁC) solution of chlorophosphite 2
31P NMR: 133.0. H NMR: 1.01 (t, 6H, JH–H = 6 Hz),
3
2.6–2.9 (m, 4H, PCH2), 2.97 (d, 3H, CH3–N, JP–H
= 10 Hz), 6.5–7.2 (m, 4H, Ar).

O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
5555
3
3.2.5. General procedure for amidophosphites 5a–e
The mixture of diamidophosphite 4 (0.45 g, 2 mmol)
and equimolar amount of the corresponding alcohol was
measured in a flask, mixed thoroughly to form a homoge-
neous solution and heated to 160 ꢁC using an oil bath.
After 15–20 min, the formation of diethylamine was com-
plete. The bath was removed and the flask was cooled to
100 ꢁC and kept at this temperature in vacuum over 1 h
to give the desired product in a purity of 99% according
NMR data.
NMR: 2.57 and 3.25 (both d, 3H, CH3N, JP–H = 10 Hz);
3
3
4.53, 4.66 (both dq, 1H, CH, JF–H = 6 Hz, JP–H
=
2 Hz); 6.35–7.72 (m, 8H, Ar). 13C NMR: 27.8 (d, NCH3,
2JP–C = 19.7 Hz); 72.5 (dq, CH, JC–F = 32.8 Hz, JP–C
=
2
2
3.7 Hz); 107.4; 111.6; 119.6; 124.5 (3JP–C = 3.2 Hz); 122.8
1
1
(q, CH–CF3, JC–F = 280.0 Hz); 123.6 (q, ArCF3, JC–F
=
271.0 Hz); 126.2 (3JP–C = 3.8 Hz), 128.8, 130.5 (q, CArCF3,
2JC–F = 32.1 Hz); 130.7, 135.5 (d, CArNMe, JP–C
=
2
2
2.9 Hz); 148.4 (d, CArOP, JP–C = 11.7 Hz). Anal. Calc.
for C16H12F6NO2P: C, 48.62, H, 3.06; N, 3.54. Found: C,
48.67; H, 3.04; N, 3.67%.
3.2.6. 3-Methyl-2-(2,2,2-trifluoro-1-phenylethoxy)-2,3-
dihydro-1,3,2-benzoxazaphosphole 5a
3.2.9. 3-Methyl-2-{[1-(trifluoromethyl)heptyl]oxy}-2,3-
dihydro-1,3,2-benzoxazaphosphole 5d
Yield 92%, 31P NMR: 129.0 (A), 131.0 (B) (both br.s,
3
A:B = 1.5:1); 19F NMR: ꢀ76.9 (appeared t, JF–H
=
Yield 95%, 31P NMR: 127.8 (A, br.s), 139.0 (B, m)
4JP–F = 5,6 Hz); 1H NMR: 2.51 and 3.21 (both d, 3H,
3
(A:B = 1.3:1); 19F NMR: ꢀ78.96 (dd, JF–H = 6.0 Hz,
3
3
4
4JP–F = 2.1 Hz), ꢀ79.37 (dd, JF–H = 7.5 Hz, JP–F = 7.1
CH3N, JP–H = 12 Hz); 4.40–4.61 (m, 1H, CH); 6.42–7.51
2
(m, 9H, Ar). 13C NMR: 27.0 (A) (d, NCH3, JP–C
=
1
3
Hz); H NMR: 0.92 (t, 3H, CH3-C, JH–H = 6.1 Hz), 1.4–
1.2 (m, 8H, –(CH2)4–); 1.57 (q, 2H, OCH2, JH–H = 6.2
Hz), 3.16 and 3.18 (both d, 3H, CH3N, JP–H = 10 Hz);
2
3
19.7 Hz), 27.6 (B) (d, NCH3, JP–C = 20.4 Hz); 73.0 (A)
2
2
3
(q, CH, JC–F = 32.8 Hz), 73.1 (B) (q, CH, JC–F = 32.8
Hz); 107.3 (B), 107.9 (A); 108.3 (B) (d, J = 4.8 Hz), 109.0
(A) (d, J = 12.1 Hz); 111.0 (A), 111.6 (B); 119.1 (A),
119.2 (B); 119.8 (A), 120.1 (B); 122.75; 123.0 (q, CF3,
1JC–F = 280.1 Hz); 123.1 (qd, CF3, 3JP–C = 2.2 Hz,
1JC–F = 280.0 Hz); 127.4, 128.1 (B); 128.3 (A), 129.3;
132.8 (B), 133.0 (A) (both s, CArCH); 135.1 (d, CArNMe
3.65–3.80 and 4.10–4.25 (both m, 1H, CH); 6.90–6.75,
7.00–7.10 (m, 4H, Ar). 13C NMR: 13.9 (CH3); 22.4 and
2
22.5; 24.37; 27.6 (d, NCH3, JP–C = 19.1 Hz), 28.0 (d,
2
2
NCH3, JP–C = 20.7 Hz); 28.0 (A) (d, NCH3, JP–C = 20.4
3
Hz); 28.5 and 28.7; 29.0 and 29.4; 31.4 (d, JP–C = 3.0
Hz); 71.7 (q, CH, JC–F = 30.6 Hz), 71.9 (q, CH, JC–F
2
2
=
2
2
(A), JP–C = 2.9 Hz), 135.7 (d, CArNMe (B), JP–C
=
31.4 Hz); 107.6, 108.0; 111.4, 111.7; 118.8, 119.0 (d,
2
1
3JP–C = 5.3 Hz); 122.6, 122.7; 124.1 (dq, CF3, JC–F
=
2.9 Hz), 148.7 (B) (d, CArOP, JP–C = 11.7 Hz), 149.6 (A)
(d, CArOP, JP–C = 11.6 Hz). Anal. Calc. for C15H13F3-
2
3
274.0 Hz, JP–C = 13.1 Hz); 135.8, 135.4 (CArNMe); 148.7
2
2
NO2P: C, 55.05, H, 4.00. Found: C, 54.70; H, 4.36%.
(d, CArOP, JP–C = 10.7 Hz), 149.1 (d, CArOP, JP–C
=
11.5 Hz). Anal. Calc. for C15H21F3NO2P: C, 53.73; H,
6.31; N 4.18. Found: C, 53.62; H, 6.31; N, 4.30%.
3.2.7. 3-Methyl-2-[2,2,2-trifluoro-1-(3-
methoxyphenyl)ethoxy]-2,3-dihydro-1,3,2-
benzoxazaphosphole 5b
Yield 90%, 31P NMR: 123.9 (A), 125.7 (B) (both br.s,
3.2.10. 3-Methyl-2-[1-(pentafluorophenyl)ethoxy]-2,3-
dihydro-1,3,2-benzoxazaphosphole 5e
A:B = 1.3:1); 19F NMR: ꢀ77.84 (appeared t, JF–H
=
3
Yield 95%, 31P NMR: 123.3 (A), 124.6 (B) (both m,
A:B = 1.15:1); 19F NMR: 143.5 (m, o-F), 156.6 (m, pꢀ
4JP–F = 7,5 Hz); 1H NMR: 2.56 and 3.21 (both d, 3H,
3
CH3N, JP–H = 8 Hz); 3.76, 3.77 (both s, 3H, ICH3);
1
4.45–4.63 (m, 1H, CH); 6.45–7.25 (m, 8H, Ar). 13C
F), 163.0 (m, m-F); H NMR: 1.50 and 1.52 (both d, 3H,
3
2
CH3, JH–H = 8 Hz); 2.90 and 3.21 (both d, 3H, CH3N,
NMR: 27.1 (A) (d, NCH3, JP–C = 19.2 Hz), 27.7 (B) (d,
3
3JP–H = 10 Hz); 5.01 (appeared quintet, 1H, CH, JH–H
=
2
NCH3, JP–C = 20.1 Hz); 55.0 (OCH3); 73.0 (q, CH,
3JP–H = 8 Hz); 6.50–7.12 (m, 4h, Ph). 13C NMR: 22.3
(CH3CH), 27.5 (B) (d, NCH3, JP–C = 19.6 Hz), 28.0 (A)
2JC–F = 32.9 Hz); 107.3 (A), 107.8 (B); 108.7 (A) (d, J =
5.2 Hz), 109.0 (B) (d, J = 9.1 Hz); 111.0 (B), 111.6 (A),
2
2
3
(d, NCH3, JP–C = 20.4 Hz); 62.4 (OCH); 107.0 (B), 107.4
112.5 (A), 112.9 (B); 115.3; 119.2 (A) (d, JP–C = 16.2
3
3
(A); 108.7 (B) (d, J = 4.6 Hz), 109.0 (A) (d, JP–C
=
Hz), 119.7 (B) (d, JP–C = 8.1 Hz), 122.7 (d, JP–C = 6.0
3
1
11.0 Hz); 110.7 (A), 111.2 (B); 118.9 (A), 119.0 (A); 119.8
(A), 120.1 (B); 122.5 (A), 122.6 (B); 149.1 (d, CArOP,
2JP–C = 11.6 Hz); 135.8 (CArNMe); 138.37 (br); 139.2
(br); 141.7 (br); 143.2 (br); 145.4 (br); 149.1 (d, CArOP,
2JP–C = 11.6 Hz). Anal. Calc. for C15H11F5NO2P: C,
49.60, H, 3.86; N, 3.86. Found: C, 49.61; H, 3.86; N, 3.97%.
Hz); 123.1 (qd, CF3, JP–C = 10.4 Hz, JC–F = 279.0 Hz);
3
135.2 (B) and 135.7 (A) (CArNMe); 134.2 (d, JP–C = 16.9
2
Hz); 148.7 (A) (d, CArOP, JP–C = 12.1 Hz); 149.5 (B) (d,
2
CArOP, JP–C
= 11.2 Hz); 159.16 (B) (s, CArOMe),
159.22(A) (s, CArOMe). EI-MS, m/z 358 (M+H)+.
3.2.8. 3-Methyl-2-{2,2,2-trifluoro-1-[3-
(trifluoromethyl)phenyl]ethoxy}-2,3-dihydro-1,3,2-
benzoxazaphosphole 5c
3.2.11. Amidophosphites 8a–c, 9a–c (general procedure)
Triethylamine (0.2 g, 2 mmol) was added to a stirred
and cooled (ꢀ5 to 0 ꢁC) solution of chlorophosphite 6
(0.28 g, 2 mmol) for compounds 8a–c or to that of chloro-
phosphite 7 (0.3 g, 2 mmol) for compounds 9a–c in 30 mL
Yield 95%, 31P NMR: 124.0 (A), 124.5 (B) (both m,
A:B = 1.3;1); 19F NMR: ꢀ63,5 (br.s, 3-CF3Ar), ꢀ78.13
3
(appeared t, CH(CF3), JF–H = 4JP–F = 7,5 Hz); 1H

5556
O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
of diethyl ether under argon and the mixture was stirred for
additional 5 min. Then the solution of 2 mmol of the corre-
sponding alcohol in 30 mL of diethyl ether was added. The
mixture was allowed to warm up to ambient temperature
and stirred for additional 3 h. The precipitate was filtered
off under argon, washed with Et2O (2 · 30 mL) and dis-
carded. The filtrate was evaporated in vacuo. The residue
was treated with 30 mL of pentane and kept at 0 ꢁC over-
night. The clear solution was decanted from the precipitate
and evaporated under reduced pressure followed by drying
(20 ꢁC, 0.1 mmHg, 2 h) to afford the desired amidophosph-
ites (8a–c, 9a–c) as a colourless oils.
(m, 2H, NCH2); 4.10–4.54 (m, OCH2, 2H); 5.35–5.50 (m,
1H, OCH). 13C NMR: 22.2 (br., CH3); 30.9 (d, CH3N,
2
2JP–C = 15.3 Hz), 31.1 (d, CH3N, JP–C = 15.4 Hz); 48.8
2
2
(d, CH2N, JP–C = 5.1 Hz), 48.9 (d, CH2N, JP–C = 5.1
2
Hz); 66.4 (d, OCH, JP–C = 5.8 Hz), 66.7 (d, OCH,
2JP–C = 5.8 Hz); 68.7 (d, OCH2, JP–C = 5.8 Hz), 130.6
2
(br), 135.8 (br.m), 136.0 (br.m), 138.5 (br.m), 143.1
(br.m), 145.7 (br.m). Anal. Calc. for C11H11F5NO2P:
C, 41.92; H, 3.52; N, 4.44. Found: C, 41.48; H, 3.65; N,
4.42%.
3.2.15. 1,3-Dimethyl-2-(2,2,2-trifluoro-1-phenylethoxy)-
1,3,2-diazaphospholidine 9a
3.2.12. 3-Methyl-2-{2,2,2-trifluoro-1-[3-
(trifluoromethyl)phenyl]ethoxy}-1,3,2-oxazaphospholidine
8a
Yield 88%. 31P NMR: 132.4 (br.s); 19F NMR: ꢀ77.11
3
4
1
(dd, JF–H = 7.0 Hz, JP–F = 4.1 Hz); H NMR: 2.46 and
3
2.57 (both d, 6H, CH3N, JP–H = 12.2 Hz); 3.00–3.35 (m,
Yield 90%. 31P NMR: 142.1 (A), 143.7 (B) (both br.s,
A:B = 1:1.3); 19F NMR: ꢀ63.5 (br.s, CF3Ar), ꢀ77,9 and
4H, CH2N); 5.12 (appeared quintet, 1H, CH, JP–H
=
3
3JP–H = 7.0 Hz,); 7.35–7.51 (m, 5H, Ph). 13C NMR: 33.0
1
2
2
ꢀ78,2 (both br.s, CF3CH, A and B); H NMR: 2.72 and
(d, CH3N, JP–C = 22.6 Hz), 33.4 (d, CH3N, JP–C =
24.8 Hz); 52.1 (d, CH2N, JP–C = 10.9 Hz), 52.5 (d,
3
2
2.64 (d, 3H, CH3N, JP–H = 12 Hz); 2.95–3.30 (m, 2H,
2
2
CH2N); 5.18–5.35 (m, 1H, CH); 3.45–4.12 (m, 2H,
CH2N, JP–C = 10.2 Hz); 72.1 (qd, CH, JC–F = 31.3 Hz,
2
1
CH2O). 13C NMR: 30.5 (d, CH3N, JP–C = 20.4 Hz), 30.6
2JP–C = 7.3 Hz); 124.1 (qd, CHCF3, JC–F = 280.1 Hz,
2
2
(d, CH3N, JP–C = 21.9 Hz); 48.4 (d, CH2N, JP–C
=
3JP–C = 3.6 Hz);127.6 (o-CAr), 128.0 (m-CAr), 128.7 (p-
CAr), 134.7 (ipso-CAr). EI-MS, m/z 293 (M+H)+.
2
5.8 Hz), 48.7 (d, CH2N, JP–C = 5.1 Hz); 69.0 (d, CH2O,
2JP–C = 9.5 Hz), 69.1 (d, CH2O, JP–C = 10.2 Hz); 71.7
2
2
2
(dq, CH, JC–F = 32.8 Hz, JP–C = 13.1 Hz), 72.1 (dq,
2
2
3.2.16. Diethyl 1-[(1,3-dimethyl-1,3,2-diazaphospholidin-2-
yl)oxy]-2,2,2-trifluoroethylphosphonate 9b
CH, JC–F = 32.8 Hz, JP–C = 15.3); 123.5 (dq, CHCF3,
1JC–F = 283.1 Hz, JP–C = 7.3 Hz); 124.3 (m); 124.8 (q,
3
Yield 96%. 31P NMR: 14.3 (m, PIV), 144.9 (br.s, PIII);
1
Ar-CF3, JC–F = 269.2 Hz); 125.9 (m); 128.75; 128.83;
130.6 (d, OCH–CAr, JP–C = 7.3 Hz); 130.9 (br); 135.2;
3
4
19F NMR: ꢀ72.11 (dd, JF–H = 8.0 Hz, JP–F = 7.5 Hz);
3
1H NMR: 1.35 (t, 6H, CH3C, JH–H = 6.1 Hz); 2.71 and
3
135.3. Anal. Calc. for C12H12F6NO2P: C, 41.51; H, 3.48;
N, 4.03. Found: C, 41.04; H, 3.53; N, 3.90%.
3
2.77 (both d, 6H, CH3N, JP–H = 12.1 Hz); 3.05–3.35 (m,
4H, CH2N); 4.11–4.40 (m, 4H, CH2O), 4.53 (dq, 1H,
3
2
CH, JF–H = 8.0 Hz, JP–H = 20.1 Hz). 13C NMR: 16.0
3
2
3.2.13. Diethyl 2,2,2-trifluoro-1-[(3-methyl-1,3,2-
oxazaphospholidin-2-yl)oxy]ethylphosphonate 8b
Yield 85%. 31P NMR: 13.5 (m, PIV), 147.2 (br.s, PIII);
(d, CH3CH2OP, JP–C = 5.3 Hz), 33.5 (d, CH3N, JP–C =
22.8 Hz), 33.8 (d, CH3N, JP–C = 24.3 Hz); 52.2 (d,
CH2N, JP–C = 9.9 Hz), 52.3 (d, CH2N, JP–C = 9.9 Hz);
2
2
2
1
2
19F NMR: ꢀ72.07 and ꢀ72.45 (both m); H NMR: 1.37
63.1 (d, P(O)OCH2, JP–C = 6.8 Hz), 63.2 (d, P(O)OCH2,
3
2
(t, 6H, CH3C, JH–H = 6 Hz), 2.76 and 2.80 (both d, 3H,
2JP–C = 6.9 Hz); 68.2 (dqd, OCH, JD(III)-C = 13.7 Hz,
3
1
NCH3, JP–H = 12 Hz); 3.00–3.25 (m, 2H, NCH2); 4.45–
2JC–F = 32.7 Hz, JP(IV)-C = 170.3 Hz); 122.9 (qd, CHCF3,
4.65 (m, CH, 1H); 4.10–4.50 (m, OCH2, 6H). 13C NMR:
1JC–F = 281.1 Hz, 3JP–C = 8.1 Hz). Anal. Calc. for
C10H21F3N2O4P2: C, 34.10; H, 6.01; P, 17.59. Found: C,
33.98; H, 6.00; P, 16.99%.
15.6 (d, CH3CH2OP, 3JP–C = 5.5 Hz), 15.7 (d, CH3CH2OP,
2
3JP–C = 5.5 Hz); 30.2 (d, CH3N, JP–C = 19.7 Hz), 30.6 (d,
2
2
CH3N, JP–C = 20.4 Hz); 47.8 (d, CH2N, JP–C = 5.8 Hz),
2
48.2 (d, CH2N, JP–C = 5.8 Hz); 62.9, 63.0, 63.2, 63.4 (all
2
3.2.17. 1,3-Dimethyl-2-{[1-(trifluoromethyl)heptyl]oxy}-
1,3,2-diazaphospholidine 9c
d, CH3CH2OP, JP–C = 6.6 Hz); 66.6 and 68.2 (both m,
OCH); 68.5, 68.7 (both d, PIIIOCH2, JP–C = 10.2 Hz),
2
Yield 98%. 31P NMR: 141.3 (br.s); 19F NMR: ꢀ78.2
1
122.3 (q of m, CHCF3, JC–F = 282.0 Hz). EI-MS, m/z
3
(appeared t, JF–H = 4JP–F = 7.5 Hz); 1H NMR: 0.90 (t,
340 (M+H)+.
3
3H,CH3C, JH–H = 6.2 Hz); 1.15–1.25 (m, 8H, –(CH2)4–);
1.60–1.75 (m, 2H, OCH2); 2.71 and 2.75 (d, 6H, CH3N,
3JP–H = 14.2 Hz); 3.00–3.15 and 3.25–3.35 (m, 4H,
CH2N); 4.00–4.25 (m, 1H, CH). 13C NMR: 13.7 (CH3);
22.3, 24.6, 28.7, 29.8, 31.5 ((CH2)5); 33.7 (d, CH3N,
3.2.14. 3-Methyl-2-[1-(pentafluorophenyl)ethoxy]-1,3,2-
oxazaphospholidine 8c
Yield 82%. 31P NMR: 141.0 and 140.7 (overlapping m);
19F NMR: 143.4 (m, o-FAr), 156.2 (m, p-FAr), 162.8 (m, m-
2
2JP–C = 21.9 Hz), 34.2 (d, CH3N, JP–C = 24.8 Hz); 52.4
1
3
2
2
FAr); H NMR: 1.61 (d, 3H, CH3CH, JH–H = 6 Hz); 2.63
(d, CH2N, JP–C = 9.5 Hz), 52.5 (d, CH2N, JP–C = 9.5
3
2
2
and 2.70 (both d, 3 H, NCH3, JP–H = 10 Hz); 2.90–3.24
Hz); 71.1 (qd, CH, JC–F = 29.9 Hz, JP–C = 17.5 Hz);

O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
5557
1
3
3
125.0 (qd, CHCF3, JC–F = 280.1 Hz, JP–C = 2.9 Hz). EI-
MS, m/z 301 (M+H)+.
Hz); 1.56 (d, 3H, CH3, JH–H = 6.0 Hz), 3.02–3.40 (m,
3
6H, CH2N + CHN); 5.39 (dq, 1H, CH, JH–H = 6.0 Hz,
3JP–H = 2.0 Hz). 13C NMR: 21.8 (d, NCHCH3, JP–H
=
3
3
8.8 Hz); 21.9 (d, NCHCH3, JP–C = 8.8 Hz); 22.0 (d,
NCHCH3, JP–C = 10.2 Hz); 22.1 (d, NCHCH3, JP–C
3.2.18. Diamidophosphites 14a–g (general procedure)
A mixture of 2 mmol of the corresponding diamine and
4 mmol of triethylamine in 30 mL of ether was added to a
cooled (0 ꢁC) solution of a dichlorophosphite (2 mmol)
13a,b in 30 mL of ether. The mixture was allowed to warm
up to room temperature and stirred for 2 h. Then the pre-
cipitate formed was filtered off, washed by an additional
amount of ether (2 · 15 mL). The combined filtrate was
evaporated to dryness, pentane was added to the residue
and kept at 0 ꢁC overnight. The clear pentane solution
was decanted, evaporated and the residue dried in vacuum
(2 h, 40 ꢁC) to yield the desired diamidophosphite. Com-
pounds 14g,f were recrystallized from hexane.
3
3
=
=
2
10.2 Hz); 22.6 (br., OCHCH3); 45.4 (d, CH2N, JP–C
2
5.1 Hz), 45.5 (d, CH2N, JP–C = 5.1 Hz); 47.4 (d, CHN,
2JP–C = 14.6 Hz); 47.7 (d, CHN, JP–C = 13.9 Hz); 61.2
2
2
(d, OCH, JP–C = 10.2); 135.8 (m); 138.6 (m); 141.6 (m);
143.1 (m); 145.6 (m). EI-MS, m/z 385 (M+H)+.
3.2.22. 1,3-Di(tert-butyl)-2-(2,2,2-trifluoro-1-
phenylethoxy)-1,3,2-diazaphospholidine 14d
Yield 84%, m.p. 47–48 ꢁC, 31P NMR: 123.9; 19F NMR:
ꢀ76,9; 1H NMR: 1.12, 1.15 (two s, 2 · 9H, CH3); 3.00–3.35
(m, 4H, CH2N); 5.14 (dq, 1H, CH, 3JF–H = 8.0 Hz, 3JP–H
=
2.0 Hz); 7.30–7.52 (m, 5H, Ph). 13C NMR: 29.1 (d,
3
3
NCCH3, JP–C = 10.9 Hz); 29.3 (d, NCCH3, JP–C = 10.9
2
3.2.19. 1,3-Diisopropyl-2-(2,2,2-trifluoro-1-phenylethoxy)-
1,3,2-diazaphospholidine 14a
Hz); 44.8 (d, CH2N, JP–C = 8.8 Hz); 45.2 (d, CH2N,
2JP–C = 8.8 Hz); 52.2 (d, NC(CH3)3, JP–C = 6.7 Hz); 52.3
2
Yield 87% (oil), 31P NMR: 129.3; 19F–{1H} NMR:
ꢀ77.8; 1H NMR: 1.02, 1.03, 1.04, 1.16 (all d, 4 · 3H,
2
2
(d, NC(CH3)3, JP–C = 3.4 Hz); 72.0 (qd, OCH, JC–F
30.6 Hz, JP–C = 6.6 Hz); 124.4 (qd, CF3, JC–F = 281.0
=
2
1
3
3
CH3 (i-Pr), JH–H = 6.0 Hz); 3.01–3.32 (m, 6H, CH2N +
Hz, JP–C = 5.1 Hz); 127.9 (o-CAr), 128.3 (m-CAr), 128.7
3
3
(p-CAr), 135.5 (ipso-CAr). EI-MS, m/z 377 (M+H)+.
CHN); 5.15 (dq, 1H, CH, JF–H = 8.0 Hz, JP–H = 2.1
Hz); 7.25–7.51 (m, 5H, Ph). 13C NMR: 21.5 (d, CH3,
3JP–C = 7.3 Hz); 21.8 (d, CH3, JP–C = 9.5 Hz); 21.9 (d,
3
3.2.23. 1,3-Di(tert-butyl)-2-[1-
(pentafluorophenyl)ethoxy]-1,3,2-diazaphospholidine 14e
Yield 86%, m.p. 43–45 ꢁC, 31P NMR: 120.0; 19F NMR:
3
3
CH3, JP–C = 8.8 Hz); 22.1 (d, CH3, JP–C = 9.5 Hz); 45.0
(d, CH2N, JP–C = 9.5 Hz), 45.2 (d, CH2N, JP–C = 9.5
2
2
2
Hz); 47.1 (d, CHN, JP–C = 10.2 Hz), 47.3 (d, CH2N,
1
ꢀ143.0 (m, o-F), ꢀ157.5 (m, p-F), ꢀ163.4 (m, m-F); H
2JP–C = 12.4 Hz); 72.3 (qd, OCH, 2JC–F = 30.6 Hz,
NMR: 1.17, 1.22 (two s, 2 · 9H, CH3); 1.56 (d, 3H, CH3,
3JH–H = 6.1 Hz); 3.25–3.40 (m, 4H, CH2N); 5.45 (dq, 1H,
2JP–C = 10.2); 124.2 (qd, CF3, JC–F = 281.0 Hz, JP–C
=
1
3
4.4 Hz); 125.7, 127.8. 128.7, 135.1 (ipso-CAr). EI-MS, m/z
349 (M+H)+. Anal. Calc. for C16H24F3N2OP: C 55.17, H
6.94. Found: C 55.41; H 6.82%.
CH, JH–H = 6.1 Hz, JP–H = 4.1 Hz). 13C NMR: 22.8
3
3
3
(br., OCHCH3); 29.1 (d, NCCH3, JP–C = 10.7 Hz); 29.3
(d, NCCH3, JP–C = 10.6 Hz); 45.0 (d, CH2N, JP–C = 9.1
3
2
2
Hz); 45.3 (d, CH2N, JP–C = 9.1 Hz); 52.0 (d, NC(CH3)3,
2
3.2.20. 1,3-Diisopropyl-2-(2,2,2-trifluoro-1-
phenylethoxy)hexahydro-1,3,2-diazaphosphorine 14b
2JP–C = 4.6 Hz); 52.2 (d, NC(CH3)3, JP–C = 3.8 Hz); 60.6
2
(d, OCH, JP–C = 8.6 Hz); 136.1 (m); 138.5 (m); 141.1 (m)
Yield 84% (oil); 31P NMR: 129.9; 19F NMR: ꢀ78.3; H
143.3 (m); 145.7 (m). EI-MS, m/z 413 (M+H)+.
1
NMR: 0.79, 1.00, 1.15, 1.20 (all d, 4 · 3H, CH3 (i-Pr),
3JH–H = 6.0 Hz); 2.51–3.5 (m, 8H, CH2CH2N + CHN); 4.92
3.2.24. 1,3-Diphenyl-2-(2,2,2-trifluoro-1-phenylethoxy)-
1,3,2-diazaphospholidine 14f
3
3
(dq, 1H, CH, JF–H = 8.0 Hz, JP–H = 2.0 Hz); 7.25–7.75
(m, 5H, Ph). 13C NMR: 20.6 (d, CH3, JP–C = 10.9 Hz);
3
Yield 83%, m.p. 119–120 ꢁC; 31P NMR: 107.2; 19F–{1H}
3
3
1
20.7 (d, CH3, JP–C = 10.9 Hz); 21.8 (d, CH3, JP–C
10.2 Hz); 22.1 (d, CH3, JP–C = 8.8 Hz); 26.8 (NCH2CH2-
=
NMR: ꢀ77.7; H NMR: 3.45–3.91 (m, 4H, CH2N); 5.04
3
3
3
(dq, 1H, CH, JF–H = 7.5 Hz, JP–H = 2.1 Hz); 6.30–7.00,
2
7.05–7.25 (m, 15H, Ph). 13C NMR: 45.6 (d, CH2N,
CH2N); 37.6 (d, CH2N, JP–C = 5.1 Hz), 38.2 (d, CH2N,
2JP–C = 5.8 Hz); 51.5 (d, CHN, JP–C = 8.0 Hz), 51.9 (d,
2
2JP–C = 10.2 Hz); 47.2 (d, CH2N, JP–C = 9.5 Hz); 72.8 (q,
2
2
2
2
1
CH2N, JP–C = 12.4 Hz); 72.8 (qd, OCH, JC–F = 31.4
OCH, JC–F = 32.1 Hz); 123.6 (qd, CF3, JC–F = 279.3
Hz, JP–C = 5.1 Hz); 127.5, 127.9, 128.9, 129.1, 132.7;
2
1
3
Hz, JP–C = 14.6); 124.4 (qd, CHCF3, JC–F = 280.7 Hz,
3JP–C = 7.3 Hz); 127.8, 127.9. 128.6, 135.2 (ipso-CAr). EI-
MS, m/z 363 (M+H)+.
2
143.9 (ipso-CArN, JP–C = 16.8 Hz); 144.2 (ipso-CArN,
2JP–C = 18.9 Hz). Anal. Calc. for C22H20F3N2OP: C,
63.46, H, 4.84; N, 6.73. Found: C, 63.49; H, 4.83; N, 6.60%.
3.2.21. 1,3-Diisopropyl-2-[1-(pentafluorophenyl)ethoxy]-
1,3,2-diazaphospholidine 14c
3.2.25. 2-[1-(Pentafluorophenyl)ethoxy]-1,3-diphenyl-
1,3,2-diazaphospholidine 14g
Yield 75%, m.p. 107–108 ꢁC, 31P NMR: 108.0; 19F
NMR: ꢀ144.0 (m, o-F), ꢀ157.0 (m, p-F), ꢀ163.0 (m,
Yield 80% (oil), 31P NMR: 125.0; 19F NMR: ꢀ143.2 (m,
1
o-F), ꢀ157.2 (m, p-F), ꢀ163.2 (m, m-F); H NMR: 1.10,
3
1.11, 1.12, 1.13 (all d, 4 · 3H, CH3 (i-Pr), JH–H = 6.0

5558
O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
1
3
Table 3
m-F); H NMR: 1.48 (d, 3H, CH3, JH–H = 6.2 Hz), 3.62–
3
Crystal data and structure refinement parameters for 14f and 15a
3.91 (m, 4H, CH2N); 5.38 (dq, 1H, CH, JH–H = 6.2 Hz,
3JP–H = 2.1 Hz); 6.25–7.02, 7.10–7.40 (m, 10H, Ph). 13C
NMR: 22.3 (OCHCH3); 62.6 (OCH); 46.8 (d, CH2N,
Compound
14f
15a
Empirical formula
Formula weight
Temperature (K)
Crystal system
Space group
C22H20F3N2OP
416.37
120
C26H27Cl2F6NO2PRh
704.27
120
Monoclinic
C2
15.052(2)
8.9471(11)
22.443(3)
2JP–C = 9.4 Hz); 114.3 (d, JP–C = 1.9 Hz); 115.6 (d,
4
4JP–C = 1.8 Hz); 119.9; 120.4; 129.1 (d, JP–C = 25.2 Hz);
3
Triclinic
129.5; 135.7 (m); 138.6 (m); 141.9 (m) 142.4 (m); 144.3
(ipso-CArN, JP–C = 16.6 Hz); 144.5 (ipso-CArN, JP–C
16.3 Hz). EI-MS, m/z 452 (M+H)+.
ꢀ
P1
2
2
=
˚
a (A)
6.3451(10)
10.0166(16)
15.513(3)
88.911(5)
88.667(5)
83.709(5)
979.6(3)
2(1)
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3.3. Synthesis of the (p-pentamethylcyclopentadienyl)-
(amidophosphite)rhodium dichlorides 15a–c (typical
procedure)
108.234(5)
3
˚
V (A )
2870.7(7)
4(1)
1416
1.630
9.01
Z (Z0)
F(000)
432
1.412
1.84
The solution of [RhCp*Cl2]2(128 mg, 0.207 mmol) in
3 mL of CH2Cl2 was added to a solution of the correspond-
ing (amido)phosphite (0.407 mmol) in 7 mL of CH2Cl2.
After stirring the immediately formed red solution for 1 h
at room temperature, the solvent was evaporated in vacuo
up to the volume of ca. 0.5 mL and then about 4 mL of
pentane was added. The crystal precipitated (dark-orange)
was filtered off and dried in vacuo to afford the desired
complexes.
Dcalc (g cmꢀ1
)
Linear absorption,
l (cmꢀ1
Tmin/Tmax
Flack parameter
Scan type
h Range (ꢁ)
Completeness of
dataset (%)
Reflections measured
Independent reflections
Observed reflections
[I > 2r(I)]
Parameters
Final R(Fhkl): R1
wR2
)
0.9572/0.9765
–
x
1.91–27.5
99.4
0.7145/0.9153
0.02(6)
x
1.91–27.50
99.3
7697
8802
4474 [Rint = 0.0317]
2849
5302 [Rint = 0.0515]
3272
Compound 15a. Yield 86%, m.p. 202–203 ꢁC, 31P NMR:
130.38 (d, JP–Rh = 229.2 Hz); 19F–{1H} NMR: ꢀ62.63 (s,
1
3F, CF3C6H4,), ꢀ76.23 (s, 3F, CF3CH); 1H NMR: 1.69 (d,
262
368
15H, CH3 in Cp*, J = 6.0 Hz); 3.12 (d, 3H, CH3N, 3JP–H
=
0.0457
0.0995
0.957
0.417, ꢀ0.283
0.0578
0.1021
0.999
1.052, ꢀ0.926
3
10.0 Hz); 5.69, 5.78 (both dq, 1H, CH, JF–H = 6.1 Hz,
Goodness-of-fit value
Dqmax, Dqmin (e A
3JP–H = 3.2 Hz); 6.71 (d, 1H, Ar, JH–H = 7.7); 6.80–6.87
3
ꢀ3
˚
)
3
(m, 2H, Ar); 7.02 (t, 1H, Ar, JH–H = 7.7 Hz); 7.35–7.45
(m, 3H, Ar); 7.60 (d, 1H, Ar, 3 JH–H = 7.2 Hz). Anal. Calc.
for C26H27Cl2F6NO2PRh: C, 44.34, H, 3.86. Found: C,
44.36; H, 3.91%.
carried out with a Bruker SMART 1000 CCD area detec-
tor, using graphite monochromated Mo Ka radiation
(kl = 0.71073 A, x-scans) at 120 K (Table 3). Low temper-
Compound 15b. Yield 89%, m.p. 223–224 ꢁC, 31P NMR:
116.48 (d, 1JP–Rh = 196.8 Hz); 19F–{1H} NMR: ꢀ76.08; 1H
NMR: 1.58 (d, 15H, CH3 in Cp*, J = 4.9 Hz); 2.39 (br.d,
˚
ature of the crystals was maintained with a Cryostream
(Oxford Cryosystems) open-flow N2 gas cryostat. Reflec-
tion intensities were integrated using SAINT software [16]
and absorption correction was applied semi-empirically
using SADABS program [17]. The structures were solved by
direct method and refined by the full-matrix least-squares
against F2 in anisotropic approximation for non-hydrogen
atoms. The analysis of difference Fourier synthesis has
revealed that one of CF3 groups in complex 15a is disor-
dered by two positions which were refined with equal occu-
pancies in anisotropic approximation. The positions of
hydrogen atoms were calculated from geometrical point
of view. Crystal data and structure refinement parameters
for 14f and 15a are given in Table 3. All calculations were
performed using the SHELXTL software [18].
The crystallographic data have been deposited with the
Cambridge Crystallographic Data Center, CCDC-614283
for 14f and CCDC-614284 for 15a. Copies of this informa-
tion may be obtained free of charge from: The Director,
CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK (fax:
+44 1223 336033; e-mail: deposit@ccdc.cam.ac.uk or
http://www.ccdc.cam.ac.uk).
3
3H, CH3N, JP–H = 10.6 Hz); 2.64 (br., 3H, CH3N);
3.24–3.38 (m, 4H, CH2N); 6.01 (br., 1H, CH); 7.30–7.32,
7.43–7.47 (both m, 3H + 2H, Ph). Anal. Calc. for
C22H31Cl2F3N2OPRh: C, 43.95, H, 5.20; N, 4.66. Found:
C, 44.12; H, 5.09; N, 4.77%.
Compound 15c. Yield 83%, m.p. 139–140 ꢁC, 31P NMR:
11.72 (s, PIV), 116.59 (d, PIII, 1JP–Rh = 218.2 Hz); 19F–{1H}
1
3
NMR: ꢀ69.3; H NMR: 1.31 (t, 6H,CH3C, JH–H = 5.2
Hz); 1.61 (d, 15H, CH3 in Cp*, J = 3.8 Hz); 2.83 and
3
2.86 (both d, 6H, CH3N, JP–H = 11.4 Hz); 3.27–3.35 and
3.42–3.57 (both m, 2H + 2H, CH2N); 4.18-4.28 (m, 4H,
CH2O), 5.17 (br., 1H, CH). Anal. Calc. for C20H36Cl2F3-
N2O4P2Rh: C, 36.33, H, 5.49; N, 4.24. Found: C, 36.17;
H, 5.56; N, 3.98%.
3.4. X-ray crystallography
Crystals of 14f and 15a suitable for X-ray diffraction
were grown up by slow evaporation of solutions of the
compounds in CH2Cl2. X-ray diffraction experiments were

O. Artyushin et al. / Journal of Organometallic Chemistry 691 (2006) 5547–5559
5559
[4] (a) A.P. Dobbs, M.R. Kimberley, J. Fluorine Chem. 118 (2002) 3;
3.5. Hydroformylation experiments
´
´
(b) I.T. Horvath, J. Rabai, Science 266 (1994) 72;
´
(c) J.J.J. Juliette, I.T. Horvath, J.A. Gladysz, Angew. Chem., Int. Ed.
Engl. 36 (1997) 1610;
In a typical experiment 1.6 mg (0.00625 mmol) of
Rh(CO)2(acac) precursor and 0.0125 mmol of ligand were
dissolved in 1.5 mL toluene, 3.125 mmol styrene was added
and the homogeneous solution was transferred under
argon into a stainless steel autoclave. The reaction vessel
was pressurized to 100 bar total pressure by CO/H2 = 1/1
and placed in an oil bath and the mixture was stirred with
a magnetic stirrer for the appropriate reaction time. The
pressure was monitored throughout the reaction. After
cooling and venting of the autoclave, the pale yellow solu-
tion was removed and analysed immediately by GC.
´
(d) J.J.J. Juliette, D. Rutherford, I.T. Horvath, J.A. Gladysz, J. Am.
Chem. Soc. 121 (1999) 2696;
(e) B. Betzemeier, P. Knochel, Angew. Chem., Int. Ed. Engl. 36
(1997) 2623.
´
[5] (a) I. Odinets, T. Kegl, E. Sharova, O. Artyushin, E. Goryunov, G.
Molchanova, K. Lyssenko, T.A. Mastryukova, G.-V. Ro¨schenthaler,
G. Keglevich, L. Kollar, J. Organomet. Chem. 690 (2005) 3456;
(b) I.L. Odinets, O.I. Artyushin, E.V. Sharova, E.I. Goryunov, D.G.
Golovanov, K.A. Lyssenko, P.V. Petrovskii, T.A. Mastryukova,
Mendeleev Commun. 13 (2003) 102.
´
[6] In our previous study dealing with phosphorylation of benzylic
alcohols by P(III) chlorides, we have found that compounds formed
from alcohols containing the donor substituents in the meta-position
of the aromatic ring are stable, while analogues with para-substituents
undergo decomposition on standing. To exclude such drawback and
to obtain stable ligands in the work cited in Ref. [5] and in the present
paper we used either non-substituted or meta-substituted fluorinated
benzylic alcohols.
[7] S.W. King, R. Natarajan, R. Bembi, T. Fife, J. Am. Chem. Soc. 114
(1992) 10715.
[8] E.I. Goryunov, O.I. Artyushin, I.B. Goryunova, I.L. Odinets, G.-V.
Ro¨schenthaler, T.A. Mastryukova, J. Fluor. Chem., in press.
[9] C.A. Tolman, Chem. Rev. 77 (1977) 31.
3.6. Computational details
Full geometry optimizations have been performed at the
density functional level of theory without any symmetry
constraints using the Gaussian 03 suite of programs [19]
and the 6-31G* basis set [20]. The stationary points were
characterized by frequency calculations in order to verify
that the minima have zero imaginary frequency. The
NBO analyses were carried out on the stationary points
using the NBO 3.1 program [10] as implemented in Gauss-
ian. For all the calculations the gradient-corrected
exchange functional developed by Perdew, Burke and Ern-
zerhof was utilized in combination with a correlation func-
tional also developed by the same authors [21] and denoted
as PBE. The initial structures for geometry optimizations
were obtained by Monte Carlo conformational analyses
using the Spartan ’04 program package [22] and the
MMFF force field. In order to determine the Tolman’s
cone angles [9] the STERIC program [23] was used.
[10] E.D. Glendening, A.E. Reed, J.E. Carpenter, F. Weinhold, NBO
Version 3.1.
[11] L.S. Zacharov, E.I. Goryunov, S.T. Ioffe, L.L. Morozov, T.M.
Scherbina, M.I. Kabachnik, Izv. AN USSR, Ser. Khim. (1976) 1834.
[12] A.M. Haas, G. Haegele, J. Fluorine Chem. 78 (1996) 75.
[13] P.C. Crofts, J.H.H. Markes, N.D. Rygon, J. Chem. Soc. (1958)
4250.
[14] D.B. Denney, D.Z. Denney, L-T. Liu, Phosphorus Sulfur 22 (1985)
71.
[15] W. Bhanthumnavin, W.G. Bentrude, J. Org. Chem. 12 (2005) 4643.
[16] SMART V5.051 and SAINT V5.00, Area detector control and integration
software, Bruker AXS Inc., Madison, WI 53719, USA, 1998.
[17] G.M. Sheldrick, SADABS, Bruker AXS Inc., Madison, WI 53719, USA,
1997.
[18] G.M. Sheldrick, SHELXTL-97, Version 5.10, Bruker AXS Inc., Mad-
ison, WI 53719, USA.
Acknowledgements
[19] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C.
Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin,
R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma,
G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G.
Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T.
Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challa-
combe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C.
Gonzalez, J.A. Pople, Gaussian 03, Revision B.05, Gaussian, Inc.,
Pittsburgh, PA, 2003.
I.O. and G.V.R. are grateful for the financial support
Russian Basic Research Foundation (grant No. 05-03-
32692) and German Science Foundation 436 RUS 113/
766/1-1. The authors thank the Hungarian Academy of
Sciences and the Hungarian Scientific Research Fund for
financial support under Grant No. OTKA F046959,
T042479 NI61591 and for the Supercomputer Center of
the National Information Infrastructure Development
(NIIF) Program.
References
[1] (a) J. Ansell, M. Wills, Chem. Soc. Rev. 31 (2002) 259;
(b) M. Dieguez, O. Pamies, C. Claver, Tetrahedron: Asymmetry 15
(2004) 2113;
[20] R. Ditchfield, W.J. Hehre, J.A. Pople, J. Chem. Phys. 54 (1971)
724.
(c) K.N. Gavrilov, O.G. Bondarev, A.I. Polosukhin, Russ. Chem.
Rev. 73 (2004) 726.
[21] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996)
3865.
[22] Spartan 04, Wavefunction Inc., Irvine, CA, 92612 USA, 2004.
[23] B.C. Taverner, J. Comput. Chem. 17 (1996) 1612.
[2] P.W.N.M. van Leeuwen, C. Claver (Eds.), Rhodium Catalyzed
Hydroformylation, Kluwer, Dordrecht, 2000.
[3] B. Breit, R. Winde, T. Mackewitz, R. Paciello, K. Harms, Chem.
Eur. J. 7 (2001) 3106.